Complementary Beta Regression Model for Fitting Bounded Data

The beta regression model is the commonly used approach for modeling data in the unit interval. However, there are in the literature some useful and interesting alternatives which often under-used. This paper proposes a novel regression model for bounded data, where the response variable is complementary beta distributed with mean and dispersion parameters. The proposed regression model is a natural strong competitor of the beta regression model. The maximum likelihood method is used for estimating the model parameters. A Monte Carlo experiment is conducted to evaluate the performances of these estimators in finite samples. The usefulness of the new regression model is illustrated by two real applications.

Automated summary

This paper proposes a novel regression model for bounded data as an alternative to the commonly used beta regression model. The new model, which utilizes maximum likelihood estimation, is shown to be a strong competitor and is evaluated through Monte Carlo experiments and real applications.